\documentclass[titlepage]{article}
\usepackage{array}
\usepackage{tabularx}
\usepackage{caption}
\usepackage{booktabs}
\usepackage{float}
\usepackage{listings, lstautogobble}
\usepackage{upquote}
\usepackage[margin=2cm]{geometry}
\usepackage{graphics}
\usepackage{multirow}
\usepackage{adjustbox}
\usepackage{tikz}
\def\checkmark{\tikz\fill[scale=0.4](0,.35) -- (.25,0) -- (1,.7) -- (.25,.15) -- cycle;}

\makeatletter
\def\ext@figure{toc}
\def\ext@table {toc}

\let\originalnumberline\numberline

\def\l@figure{\def\numberline##1{\makebox[2cm][l]{Figure\hrulefill##1: }%
                                \global\let\numberline\originalnumberline}%
              \l@subsection}

\def\l@table{\def\numberline##1{\makebox[2cm][l]{Table\hrulefill##1: }%
                                \global\let\numberline\originalnumberline}%
             \l@subsection}
\makeatother

\lstset{autogobble=true}



\begin{document}


 \setcounter{figure}{0}
\renewcommand{\thefigure}{\arabic{figure}}

\setcounter{page}{35}

\begin{figure}[H]
 \caption{States with Paid Sick Leave Laws, 2021}
 \hspace{-2cm}
\resizebox{1.2\textwidth}{!}{
\includegraphics{exhibits/Figure 1.png}
}
\hspace{-0.4cm} \begin{minipage}{\textwidth}
\textit{Notes:} In March of 2021, the PSL mandates in Austin, Dallas, and San Antonio, Texas, were deemed to be preempted by a state law and consequently were struck down. \vspace{0.25cm} \\
\textit{Source}: ``Interactive Overview of Paid Sick Time Laws in the United States,'' A Better Balance. https://www.abetterbalance.org/paid-sick-time-laws/.
\end{minipage}
\end{figure}

\newpage

\begin{figure}[H]
\caption{Percentage of a Firm’s Workers Covered by State or County PSL Mandates, 2020 by Firm}
\begin{center}
\resizebox{1.05\textwidth}{!}{
\includegraphics{exhibits/Figure 2.pdf}
}
\end{center}
\end{figure}
\hspace{-0.4cm} \begin{minipage}{\textwidth}
\textit{Notes:} Each bar represents one firm. Shares of workers in PSL mandate places are calculated at the firm-county-month level, then averaged at the year level (2020 only). \vspace{0.25cm} \\
\textit{Source:} National Partnership for Women and Families (PSL mandate data) \& Reference USA (employee/establishment data), 2020.
\end{minipage}
\clearpage
\newpage


\begin{figure}[H]
 \caption{Predicted Share of Workers in Non-PSL Mandate Places Reporting Access to Paid Sick, by Firm Exposure to PSL Mandates}
\centering
\hspace{-0.95cm}
\resizebox{1.05\textwidth}{!}{
\includegraphics{exhibits/Figure 3.pdf}
}
\end{figure}
\hspace{-0.4cm} \begin{minipage}{\textwidth}
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a continuous regressor. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{table}[H]
\caption{Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places}
\resizebox{1.05\textwidth}{!}{
\input{exhibits/Table 1.tex}
}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth}
\textit{Notes:} Models regress reported access to PSL among workers in states or counties without PSL mandates on the percent of workers/establishments at the same firm who are in states or counties with PSL mandates. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}



\newpage

\begin{table}[H]
\caption{Moderation of Association between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-Mandate Places \textemdash  Firm Headquarters Location}
% \begin{adjustwidth}{-2cm}{}
\centering
\resizebox{1.05\textwidth}{!}{
\input{exhibits/Table 2.tex}
}
\begin{minipage}{\textwidth} \vspace{0.5cm} 
\textit{Notes:} Dependent variable: reported access to PSL among workers in states/counties not covered by PSL mandates. All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}
\end{table}

% \end{adjustwidth}

\newpage

\begin{table}[H]
\caption{Moderation of Association between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places \textemdash  Firm Franchising}
\centering
\input{exhibits/Table 3.tex}
\end{table}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Dependent variable: reported access to PSL among workers in states/counties not covered by PSL mandates. ``Franchisor firm'' = binary indicator. All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}



\newpage

\clearpage


 \setcounter{table}{0}
\renewcommand{\thetable}{A\arabic{table}}

 \setcounter{figure}{0}
\renewcommand{\thefigure}{A\arabic{figure}}


\begin{figure}[H]
 \caption{Share of Workers in Non-PSL Mandate Places Reporting Access to Fringe Benefits, by Firm Exposure to PSL Mandates}
 \vspace{-0.2cm}
 \hspace{-1.75cm}
\resizebox{1.2\textwidth}{!}{
\includegraphics{exhibits/Appendix Figure 1.pdf}
}
\end{figure}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a continuous regressor. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{figure}[H]
 \caption{Predicted Share of Workers in Non-PSL Mandate Places Reporting Access to Paid Sick by Firm Exposure to PSL Mandates, from Non-Linear Specification}
\begin{center}
\ \\
\includegraphics{exhibits/Appendix Figure 2.pdf}
\end{center}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a categorical regressor. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}
\end{figure}

\clearpage
\newpage

\begin{figure}[H]
 \caption{Moderation of Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places \textemdash  Firm Headquarters Location}
\begin{center}
\ \\
\resizebox{1.05\textwidth}{!}{
\includegraphics{exhibits/Appendix Figure 3.pdf}
}
\end{center}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a continuous regressor in interaction with a binary indicator of whether the firm's headquarters are exposed to a PSL mandate. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}
\end{figure}

\clearpage
\newpage

\begin{figure}[H]
 \caption{Moderation of Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places \textemdash  Firm Franchising}
\hspace{-1.7cm}
\resizebox{1.2\textwidth}{!}{
\includegraphics{exhibits/Appendix Figure 4.pdf}
}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a continuous regressor in interaction with a binary indicator of whether the firm franchises or not. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}
\end{figure}

\clearpage
\newpage


\begin{table}[H]
\centering
\caption{Robustness of Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places \textemdash  Excluding 2020/2021 \& ACS Weighting}
\resizebox{0.95\textwidth}{!}{
\input{exhibits/Appendix Table 1.tex}
}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth} 
\textit{Notes:} M2 excludes observations from 2020 \& 2021 from analytic sample. M3 employs American Community Survey weighting on the basis of age, gender, and race. All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level.  \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{table}[H]
\caption{Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places}
\hspace{-1.1cm}
\resizebox{1.15\textwidth}{!}{
\input{exhibits/Appendix Table 2.tex}
}

\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth}
\textit{Notes:} Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{table}[H]
\caption{Robustness of Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-Mandate Places to Alternative Treatment of PTO States}
\begin{centering}
\begin{adjustbox}{max width=\textwidth}
 \input{exhibits/Appendix Table 3.tex}
\end{adjustbox}
\end{centering}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth} 
\textit{Notes:} M1 \& M4 treat states with PTO mandates (ME/MI/NV) as PSL mandate states. M2 \& M5 treat states with PTO mandates (ME/MI/NV) as non-PSL mandate states. M3 \& M6 omit PTO states (ME/MI/NV) from model. All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{table}[H]
\caption{Share of Workers in Non-PSL Mandate Places Reporting Access to Fringe Benefits, by Firm Exposure to Mandates}
\centering
\begin{centering}
% \begin{adjustwidth}{-2cm}{}
\resizebox{1.05\textwidth}{!}{
\input{exhibits/Appendix Table 4.tex}
}
\end{centering}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth} 
\textit{Notes:} All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage

\begin{table}[H]
\newpage
\caption{Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places, from Non-Linear Specification}
\vspace{4mm}
\centering
\begin{centering}
\begin{adjustbox}{max width=\textwidth}
\input{exhibits/Appendix Table 5.tex}
\end{adjustbox}
\end{centering}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Reference category for share of workers covered by PSL mandate: 0-15\%. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage


\begin{table}[H]
\caption{Descriptive Statistics}
\centering
\begin{centering}
% \begin{adjustwidth}{-2cm}{}
\resizebox{0.6\textwidth}{!}{
\input{exhibits/Appendix Table 6.tex}
}
\end{centering}
\end{table}
\hspace{-0.4cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Race/ethnicity categories are mutually exclusive; i.e., respondents selecting Hispanic in addition to other races/ethnicities are classified as Hispanic. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\clearpage
\newpage


\begin{table}[H]
\caption{Moderation of Association between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-Mandate Places \textemdash  Firm Headquarters Location, Excluding IKEA}
\centering
\resizebox{1.05\textwidth}{!}{
\input{exhibits/Appendix Table 7.tex}
}
\begin{minipage}{\textwidth} \vspace{0.5cm} 
\textit{Notes:} Dependent variable: reported access to PSL among workers in states/counties not covered by PSL mandates. Excludes all observations from workers at IKEA. All models include demographic controls, work controls, firm controls, state controls, state fixed effects, and year/month/year*month fixed effects. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Firm controls: industry, total number of employees (by month), total number of establishments (by month). State controls: unemployment rate, minimum wage, union membership, breakdowns of age/gender/race/education. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}
\end{table}

\clearpage
\newpage

\begin{figure}[H]
 \caption{Moderation of Association Between Firm-Level Exposure to PSL Mandates and Workers’ Reports of Access to PSL in Non-PSL Mandate Places \textemdash  Firm Headquarters Location, Excluding IKEA}
\centering
\resizebox{0.95\textwidth}{!}{
\includegraphics{exhibits/Appendix Figure 5.pdf}
}
\end{figure}
\hspace{-0.5cm} \begin{minipage}{\textwidth} 
\textit{Notes:} Predicted values from linear regression model, with the percentage of firm's workers in a PSL mandate place as a continuous regressor in interaction with a binary indicator of whether the firm's headquarters are exposed to a PSL mandate. Excludes all observations from workers at IKEA. Demographic controls: gender, race, age, parental status, speaking a non-English language at home, being enrolled in school, cohabitation status. Work controls: hourly wage, usual weekly hours, manager status, union membership, job tenure. Also includes state and year/month/year*month fixed effects. Robust standard errors clustered at state level. \vspace{0.25cm} \\
\textit{Source:} Shift Project data, 2018-2021.
\end{minipage}

\end{document}